Paired dosimetric comparison of VMAT-based total body and total marrow irradiation in adult leukemia patients: enhanced organ sparing with consistent plan complexity.
Authors
Affiliations (2)
Affiliations (2)
- Faculty of Engineering and Natural Sciences, Fenerbahce University, Department of Electrical and Electronics Engineering, Istanbul, Türkiye.
- Department of Radiation Oncology, Medicana Bursa Hospital, Bursa, Türkiye.
Abstract
Total body irradiation (TBI) is widely used in conditioning regimens prior to hematopoietic stem cell transplantation, but it is associated with significant radiation exposure to normal tissues. Total marrow irradiation (TMI) has emerged as a more targeted alternative, aiming to reduce organ-at-risk doses while maintaining target coverage. This study aimed to evaluate whether TMI can provide clinically meaningful organ sparing while maintaining acceptable treatment complexity using a paired dosimetric approach. Thirty adult patients with acute leukemia who previously received VMAT-based TBI were retrospectively included. For each patient, a corresponding VMAT-based TMI plan was generated using the same CT datasets. To ensure planning efficiency and standardization, target delineation for TMI was facilitated by AI-based auto-segmentation. Plan quality was assessed using the homogeneity index, and complexity was analyzed based on monitor units, segment number, and modulation factor. Paired comparisons were performed using the Wilcoxon signed-rank test. TMI plans demonstrated a consistent and statistically significant reduction in dose to all evaluated organs at risk compared with TBI (p < 0.001). Mean doses to the heart, kidneys, and liver were reduced by approximately 5-6 Gy, and the lung dose was also significantly decreased. TMI provided improved dose homogeneity and showed significantly lower monitor units, segment number, and modulation factor compared with TBI. VMAT-based TMI reduced organ-at-risk doses while preserving target coverage and showing favorable plan quality and complexity metrics. AI-based auto-segmentation may reduce the workload of skeletal target delineation; however, treatment uncertainties related to patient positioning, respiratory motion, and field-junction dose matching should be carefully managed during clinical implementation.